Hypothesis Testing

Introduction
Data Observation and interpretation is a very crucial step in Data Analysis, but so is verifying the authenticity and validity of data. This target is accomplished with the help of numerous tests. One of such statistical tests is Hypothesis Testing.

Hypothesis refers to making calculative guesses about anything, considering the fact that it is testable by any means, either through experimentation or observation. A Hypothesis statement is a statement, which states a fact. That fact is to be verified via a process called Hypothesis Testing.

A Null Hypothesis is the exact opposite of the guess made. It is denoted by H0. It can be either accepted or rejected.

A simple theoretical example of the same would be:

Mr.X will wake up at 5:30 am in the morning.

Now, this is a hypothesis since it is a guess made after observing the routine of Mr.X.

The hypothesis statement is that he will wake up at exactly 5:30 am.

The Null Hypothesis here would be that Mr.X will not wake up at 5:30 am.

Mathematical representation of the same would be:

H0= Mr.X will not wake up at 5:30 am.

H1= Mr.X will wake up at 5:30 am.

This Hypothesis could be verified through observation.

If it is observed that Mr.X wakes up at 5:30 am, we will reject the null hypothesis. If it is observed that M.X wakes up before or after 5:30 am, then we will accept the Null Hypothesis. Here, H1 is called the Alternate Hypothesis.

The real-world applicability of Hypothesis Testing is far more mathematical. Examples could be testing the Null Hypothesis that the mean of some given data is more than 0.65 against the Alternate Hypothesis that the mean would be less than or equal to 0.65. i.e.

H0= 𝜇 >0.65

H1= 𝜇 <= 0.65

That’s all on Hypothesis Testing for this blog. The applicability of R in Hypothesis Testing will be covered in the next blog. Suggestions and reviews to help me improve are most welcomed. Thank You for visiting.

Shreya Kashyap